We live in an environment that often signals the likelihood of the occurrence of some future event. If stimuli are to signal future events they must be remembered. Because neuronal circuits that mediate such memories are often poorly characterized, the mnemonic processes that are necessary for past stimuli to influence future behavior have not been subjected to systematic investigation at the neuronal level. We have developed a preparation in which such studies are possible. Specifically, we will study neuronal mechanisms that mediate a preparatory state. This state is induced by signaling stimuli, and functions to improve an animal's responsiveness to subsequent stimuli (i.e., stimuli that are delivered after the signaling stimulus is no longer present). Our studies will center on the control of biting responses. We propose to test the hypothesis that the preparatory state that we study is not a single indivisible unit. Instead it consists of multiple components that can be independently controlled. We suggest that higher order neurons (e.g., CPR) activate several modulatory neurons that in turn control different components of the preparatory state. Furthermore, we suggest that the full complement of behavioral effects (including the persistence of the preparatory state) may not emerge from a simple summation of different effects. Instead behavioral effects emerge from complex interactions between different components. Preparatory states play a critical role in assuring effective actions in complex environments. Impairment in the induction or persistence of such a state is likely to give rise to serious behavioral pathologies that may manifest themselves in such disorders as depression and adult attention deficit syndrome.